test-gap-finder¶
Identify test-coverage gaps from coverage output or a static source-to-test map — distinguishing uncovered functions/paths from missing edge-case tests (untested error branches, boundaries, null/empty inputs). It reviews the given scope read-only (no test is written, no tool installed) and reports gaps grounded in the code's own branching.
At a glance¶
| Model | inherits the session's |
| Tools | Read Grep Glob Bash |
| Author | navjyotnishant |
| Source | agents/test-gap-finder.md |
Returns content — does not write files
Like all but two of the agents, this one returns its output to the skill that spawned it, and the skill writes the file. It holds no write tools, so it cannot modify the repo even if asked to.
When it runs¶
The test-gap-finder skill ran the repo's coverage tool and has an uncovered-lines report for the changed set.
what's not tested in my changes?
The skill spawns this agent with the coverage output; it separates uncovered paths from missing edge-case tests and reports, writing no tests itself.
What it returns¶
- A gaps list:
file:line/function · gap type (uncovered path / missing edge case) · the specific branch/condition untested · priority. - A one-line summary: N uncovered, M missing-edge-case, over
<scope>, via<measured coverage | heuristic map>— always name which, so the reader knows how much to trust it. - Never a written test — you describe the gap; the human (or a test-writing tool) fills it. You may suggest what a test should assert, in prose.
Spawned by¶
Derived, not declared
No agent file records which skills call it — this list is recovered from the skill definitions at build time, so it cannot go stale.
See the /test-gap-finder pipeline for where this fits in the whole run.